RESEARCH ARTICLE Open Access

Single-cell morphological characterizationof CRH neurons throughout the wholemouse brainYu Wang1, Pu Hu1, Qinghong Shan1, Chuan Huang1, Zhaohuan Huang1, Peng Chen1, Anan Li2,3, Hui Gong2,3* andJiang-Ning Zhou1,2*

Abstract

Background: Corticotropin-releasing hormone (CRH) is an important neuromodulator that is widely distributed inthe brain and plays a key role in mediating stress responses and autonomic functions. While the distributionpattern of fluorescently labeled CRH-expressing neurons has been studied in different transgenic mouse lines, a fullappreciation of the broad diversity of this population and local neural connectivity can only come from integrationof single-cell morphological information as a defining feature. However, the morphologies of single CRH neuronsand the local circuits formed by these neurons have not been acquired at brain-wide and dendritic-scale levels.

Results: We screened the EYFP-expressing CRH-IRES-Cre;Ai32 mouse line to reveal the morphologies of individualCRH neurons throughout the whole mouse brain by using a fluorescence micro-optical sectioning tomography(fMOST) system. Diverse dendritic morphologies and projection fibers of CRH neurons were found in various brainregions. Follow-up reconstructions showed that hypothalamic CRH neurons had the smallest somatic volumes andsimplest dendritic branches and that CRH neurons in several brain regions shared a common bipolar morphology.Further investigations of local CRH neurons in the medial prefrontal cortex unveiled somatic depth-dependentmorphologies of CRH neurons that exhibited three types of mutual connections: basal dendrites (upper layer) withapical dendrites (layer 3); dendritic-somatic connections (in layer 2/3); and dendritic-dendritic connections (in layer4). Moreover, hypothalamic CRH neurons were classified into two types according to their somatic locations andcharacteristics of dendritic varicosities. Rostral-projecting CRH neurons in the anterior parvicellular area had fewerand smaller dendritic varicosities, whereas CRH neurons in the periventricular area had more and larger varicositiesthat were present within dendrites projecting to the third ventricle. Arborization-dependent dendritic spines of CRHneurons were detected, among which the most sophisticated types were found in the amygdala and the simplesttypes were found in the hypothalamus.

(Continued on next page)

© The Author(s). 2021 Open Access This article is licensed under a Creative Commons Attribution 4.0 International License,which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you giveappropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate ifchanges were made. The images or other third party material in this article are included in the article's Creative Commonslicence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commonslicence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtainpermission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to thedata made available in this article, unless otherwise stated in a credit line to the data.

* Correspondence: huigong@mail.hust.edu.cn; jnzhou@ustc.edu.cn2Center for Excellence in Brain Science and Intelligence Technology, ChineseAcademy of Sciences, Shanghai 200031, China1Chinese Academy of Science Key Laboratory of Brain Function and Diseases,School of Life Sciences, Division of Life Sciences and Medicine, University ofScience and Technology of China, Hefei 230026, ChinaFull list of author information is available at the end of the article

Wang et al. BMC Biology (2021) 19:47 https://doi.org/10.1186/s12915-021-00973-x

(Continued from previous page)

Conclusions: By using the CRH-IRES-Cre;Ai32 mouse line and fMOST imaging, we obtained region-specificmorphological distributions of CRH neurons at the dendrite level in the whole mouse brain. Taken together, ourfindings provide comprehensive brain-wide morphological information of stress-related CRH neurons and mayfacilitate further studies of the CRH neuronal system.

Keywords: Corticotropin-releasing hormone, fMOST imaging, Dendritic morphology, Three-dimensionalreconstruction, Local circuit, Dendritic varicosities, Dendritic spine

BackgroundCorticotropin-releasing hormone (CRH), a 41-amino-acid peptide, is an important neuromodulator that iswidely distributed in the brain [1]. As a neuroendocrinehormone, CRH is abundantly expressed in hypothalamicparaventricular nucleus (PVN) neurons and plays a cru-cial role in the regulation of the hypothalamic-pituitary-adrenal (HPA) axis [2, 3]. CRH-expressing neurons arealso broadly distributed in other brain regions, includingthe inferior olivary nucleus, Barrington’s nucleus, pon-tine tegmentum, cerebral cortex, hippocampus, and cen-tral amygdala. Depending on their region-specificsomatic locations, CRH neurons participate in variousfunctional activities, such as learning memory, synapticplasticity, food intake, and drug addiction, as well asanxiety-like and depression-like behaviors [4–7].The anatomy of the brain CRH system has been stud-

ied in different mammalian species via immunohisto-chemistry and radioimmunoassays [8–14]. However,data from these studies have mainly been acquired fromhistological imaging and through manual reconstructionand counting of labeled neurons, which is time-consuming, limits further systematic analysis, and canintroduce biases and/or artifacts. Recently, geneticallymodified mouse models have been developed to identifythe whole-brain distributions of CRH neurons [15–19],which has significantly advanced our understanding ofthe morphological features of CRH neurons in the ro-dent brain [20–23]. Advances in whole-brain optical im-aging techniques, such as fluorescence micro-opticalsectioning tomography (fMOST) [24–27], have made itfeasible to further quantify cellular distributions and tomorphologically reconstruct cells at the whole-brainlevel. The precision of imaging via fMOST can revealcomplex fiber orientations and can even distinguish indi-vidual dendrites. Such quantitative three-dimensional(3D) neuronal morphologies obtained at a brain-widescale can provide highly accurate arborization detailsand comprehensive mapping of CRH neuronal connec-tions throughout the brain.Although the whole-brain expression patterns of CRH

have been qualitatively analyzed [17, 18, 28], high-resolution reconstruction of the full morphologies (in-cluding both somata and dendrites) of CRH neurons at

the single-neuron level has rarely been performed. Sinceneuronal morphology is considered to be one of themost defining features to distinguish among neuronaltypes and network connectivities, characterizing single-neuron morphologies may provide key information onhow neuronal information and signals are transmittedwithin the local networks. Furthermore, analysis of de-tailed morphological information (including data sets ofsomatic locations, as well as dendritic and axonal mor-phological features) of diverse CRH neurons may facili-tate a better classification of CRH neuronal types andhelp to reveal their local connectivity. For example, a re-cent study employed fMOST to investigate CRH distri-butions in the mouse brain, enabling quantitativeanalysis of whole-brain CRH somata [18] that has pro-vided us with substantial quantitative information onbrain CRH networks. However, at present, the morpho-logical details of individual CRH neuronal fibers at thewhole-brain level remain poorly understood. In thepresent study, we constructed a comprehensive whole-brain map of genetically labeled CRH neurons in themouse brain, which provides dendritic distribution pat-terns of single CRH neurons. Reconstructions and fur-ther analysis showed that heterogeneous CRHinterneurons in the mPFC form layer-dependentdendritic-dendritic and dendritic-somatic connections;furthermore, there was a target-oriented distribution ofvaricosities within the dendrites of hypothalamic CRHneurons. This work provides a comprehensive descrip-tion of the whole-brain CRH neuronal distribution pat-tern and, more importantly, dendritic morphologicalfeatures of CRH neurons in the mouse brain.

ResultsComparison of morphological features of fluorescent-labeled CRH neurons in three fluorescent-reporter mouselinesBy crossing CRH-IRES-Cre mice with Ai6, Ai14, andAi32 reporter mice, in which the cassette containingZsGreen1, td-Tomato, or CHR2-EYFP was expressed ina Cre-dependent manner, we obtained CRH-IRES-Cre;Ai6, CRH-IRES-Cre;Ai14, and CRH-IRES-Cre;Ai32 mice,respectively (Additional file 1, Figure S1. A). Then, wecompared the distributions and morphologies of

Wang et al. BMC Biology (2021) 19:47 Page 2 of 18

fluorescent-labeled CRH neurons in several brain re-gions, including the olfactory bulb (OB) (Fig. 1A, C),cortex (Fig. 1D, F), PVN (Fig. 1G, I), bed nucleus of thestria terminalis (BST) (Additional file 1, Figure S1. B andD), and central nucleus of the amygdala (CeA) (Add-itional file 1, Figure S1. E and G).In the OB, the transgenic fluorescent proteins were

mainly distributed in the glomerular layer (Gl), externalplexiform layer (EPl), and the mitral cell layer (Mi) in allthree mouse lines. CRH-IRES-Cre;Ai6 mice showed thebrightest and largest number of fluorescent-labeled cellsin all these layers (Fig. 1A); in particular, more fluores-cent cells were labeled in the granule cell layer (GrO) inthis mouse line compared to those in CRH-IRES-Cre;Ai14 and CRH-IRES-Cre;Ai32 mice. However,fluorescent-labeled neuronal fibers were short and theirfluorescent distributions were not uniform; for example,the dendrites close to the cell bodies of mitral cells werestrongly labeled (Fig. 1A, a’, indicated by the arrow-heads), but the branches extending to the Gl were notclear (Fig. 1A, a, indicated by the dotted box). CRH-IRES-Cre;Ai14 mice were also labeled with bright cellbodies and dense fibers were labeled in the EPl (Fig. 1B,b’), but only a few dendritic structures were labeled inthe Gl (Fig. 1B, b, indicated by the dotted box). By con-trast, in each layer of CRH-IRES-Cre;Ai32 mice, thefluorescence distributed in cell bodies and fibers exhib-ited a uniform brightness (Fig. 1C), and the somata inthese sections were organized in a ring-like structure(Fig. 1C, c’, indicated by the arrowhead). Unlike theformer two mouse lines, the Gl showed a bushy spher-ical structure (Fig. 1C, c) that was comprised of mitralcells and/or peribulbar cells.In the cortex, fluorescent-labeled cells were found in

each layer in CRH-IRES-Cre;Ai6 mice (Fig. 1D). The cellbodies were strongly labeled, while fibers were rarelyseen (Fig. 1D, d). The numbers of labeled cells in themedial prefrontal cortex (mPFC) of CRH-IRES-Cre;Ai14(103.4 ± 7.9/mm2) and CRH-IRES-Cre;Ai32 (108.7 ± 5.1/mm2) mice were less than those in CRH-IRES-Cre;Ai6(283.3 ± 27.8/mm2) mice (Fig. 1J, one-way ANOVA, P <0.0001, F (2, 9) = 36.56), and the cells were mainlydistributed in layer 2/3 (Fig. 1E, F). Neurons in CRH-IRES-Cre;Ai14 mice also showed clearer and brightercell bodies (Fig. 1E, e), whereas more fibers were labeled(Fig. 1F) in CRH-IRES-Cre;Ai32 mice, especially in termsof a dense distribution in the first layer (Fig. 1F, f).The outlines of nuclei were clearly visible in the fluor-

escent labeling of CRH neurons in the PVN (Fig. 1G, I,indicated by the dotted line), BST (Additional file 1, Fig-ure S1. B and D, indicated by the dotted box), and CeA(Additional file 1, Figure S1. E and G, indicated by thedotted box) of the three mouse lines. Similarly, CRH-IRES-Cre;Ai6 mice showed the highest number of

labeled neurons in the PVN (1702 ± 238.9/mm2, one-way ANOVA, P = 0.2606, F (2, 9) = 1.57), BST (714.6 ±125.3/mm2, one-way ANOVA, P < 0.005, F (2, 9) =11.38), and CeA (1085 ± 67.35/mm2, one-way ANOVA,P = 0.0778, F (2, 9) = 3.44) (Fig. 1J), as well as the stron-gest fluorescent labeling within these cell bodies (Fig. 1G;Additional file 1, Figure S1. B, a and E, d). In CRH-IRES-Cre;Ai14 mice, distinguishable cell bodies anddense fibers were labeled in the BST (Additional file 1,Figure S1. C, b) and CeA (Additional file 1, Figure S1. F,e), while only the cell bodies were clearly seen in thePVN (Fig. 1H, indicated by the dotted line). By contrast,CRH-IRES-Cre;Ai32 mice showed dense fibers in all ofthese regions, and the fluorescent signals of the cell bod-ies were distinguishable (Fig. 1I, Additional file 1, FigureS1. D, c and G, f); especially in the PVN, neuronal fibersextending to the lateral and third ventricle (Fig. 1I, i, fi-bers indicated by the arrowheads) were visible, and therewas a uniform fluorescent intensity distributed in thenearby cell bodies and fibers.In summary, among the three reporter mouse lines,

CRH-IRES-Cre;Ai6 and CRH-IRES-Cre;Ai14 miceshowed clearer and brighter cell bodies of CRH neurons.CRH-IRES-Cre;Ai6 mice had the largest number of la-beled CRH cells in each tested brain region, but almostno neuronal fibers were visible. CRH-IRES-Cre;Ai14mice showed clear but incomplete fibers. Only CRH-IRES-Cre;Ai32 mice showed the most complete fibrousstructures, especially in terms of distributions in neur-onal terminals (e.g., the bushy spherical structures in theglomerular layer of the OB; the extended fibers in thecortex and PVN); regardless of their weaknesses in dis-tinguishing single-cell bodies, the fluorescent distribu-tions in the whole cell were uniform, which is conduciveto the adjustment of exposure and the collection ofcomplete morphologies of neurons during imaging.

Whole-brain distributions of CRH neurons at highresolution in the CRH-IRES-Cre;Ai32 mouse lineSince the single-cell morphology of CRH neurons wasmost clearly visible in CRH-IRES-Cre;Ai32 mice, weused this mouse line to image EYFP-labeled CRH neu-rons throughout the brain at a resolution of 0.2 × 0.2 ×1.0 μm via an fMOST system. First, 100-μm down-sampled coronal projection sections (Fig. 2a) were pro-vided to show the overall distributions of CRH neuronsin various brain regions. EYFP-labeled cells were distrib-uted in many regions that have not previously been re-ported, such as in vascular organ of the laminaterminalis (VOLT), ventromedial preoptic nucleus(VMPO), caudate putamen (CPu), bed nucleus of the an-terior commissure (BAC), triangular septal nucleus (TS),suprachiasmatic nucleus (SCN), Kölliker-Fuse nucleus(KF), and nucleus X (X) (Fig. 2a). We analyzed the co-

Wang et al. BMC Biology (2021) 19:47 Page 3 of 18

CRH-IRES-Cre;Ai6 CRH-IRES-Cre;Ai14 CRH-IRES-Cre;Ai32

AO

BC

orte

xP

VN

ab

c

GlEPl

GrO

Mi

B Cc

a’a’

b’

b’

c’

c’

L1

L2/3

L4

L5

D E Fd

d

e

e

f

f

PVN

G H I

3V

i

i

mPFC BST PVN CeA0

250

500

750

1000

1250

1500

1750

2000

Den

sity

offlu

ores

cent

lyla

bele

dne

uron

s(c

ells

/mm

2 )

Ai6

Ai14

Ai32

J

Fig. 1 Comparison of the morphological features of CRH neurons in three fluorescent-reporter mouse lines. Comparisons of the distributions andmorphologies of fluorescent-labeled CRH neurons in several brain regions of CRH-IRES-Cre;Ai6, CRH-IRES-Cre;Ai14, and CRH-IRES-Cre;Ai32 mice inthe OB (A–C), cortex (D–F), and PVN (G–I). (A–C, a–c) The dotted boxes show ZsGreen1 (a), td-Tomato (b), and EYFP (c) labeling of synapticglobular structures in the Gl of the three mouse lines. A–B (a’–b’) The dotted boxes and magnified images show ZsGreen1 (a’) and td-Tomato(b’) labeling of cell bodies and dendrites (indicated by the arrowheads) in the Mi and EPl of CRH-IRES-Cre;Ai6 and CRH-IRES-Cre;Ai14 mice. C (c’)The dotted box and magnified image show EYFP-labeled cell body (indicated by the arrowhead) and dendrites in a CRH-IRES-Cre;Ai32 mouse.D–F (d–f) The dotted boxes and magnified images show ZsGreen1 (d), td-Tomato (e), and EYFP (f) labeling of cell bodies and dendrites in layer 1and layer 2/3 of the three mouse lines. G ZsGreen1-labeled cell bodies in (dotted-curved box) and outside (arrowheads) of the PVN of CRH-IRES-Cre;Ai6 mice. H Td-Tomato-labeled cell bodies in the PVN (dotted-curved box) of CRH-IRES-Cre;Ai14 mice. I (i) EYFP-labeled cells (I, dotted-curvedbox) and fibers (i, dotted box and magnified image) in the PVN of CRH-IRES-Cre;Ai32 mice. Scale bars = 100 μm. J The difference in density offluorescently labeled neurons in several brain regions of three reporter mouse lines. Data are shown as mean ± SEM; n = 4 mice fromindependent experiments, 4 slices per mouse

Wang et al. BMC Biology (2021) 19:47 Page 4 of 18

localization of EYFP expression with CRH immunoreac-tivity in these brain regions (Additional file 1, Figure S2.A) and found that most of the EYFP-labeled “novel”CRH neurons coexisted with CRH-immunoreactive cellsin all of the above brain regions (Additional file 1, Figure

S2. A, indicated by arrows), but low ratio of co-labelingwas observed in some regions such as in the BAC, CPu,SCN, and TS, and some CRH-immunoreactive neuronsdid not express EYFP. Furthermore, bundles of CRHprojection fibers were visible in accumbens nucleus,

OB

GrO

Mi

EPl

Gl

Cg1

DTT

Pir

M2

VMPO

cc

LS

BST

CPuBAC

SCN

TS

BLA

PVN

ME

LH

CA1DG

MM

DLG

VLG

APTD

SNC

SuG

CA1

Pn

RtTgMnR RtTg

Pr5

KF

VCA

IC

PRVCP

DC DC

MVeX

IO

ECu

IO

A

B

C D E

F G H I J

IPAC

acp icp

SCNBAC

DC

Cerebellum

VOLT ME3V

1 mm

PrLIL

AcbSh

VOLT

acp

M1S1

S2

AM

ADLD

MD

CeA

S1

DMVMH

RSG

V2V1

PoPF

AuDAu1

MGV

VLTg

icp

icp

cc

IPAC

AcbSh

CPu

Fig. 2 Whole-brain distributions of CRH neurons at high resolution in the CRH-IRES-Cre;Ai32 mouse line. a Whole-brain distributions of EYFP-labeled CRH neurons, using the fMOST system, contained in 100-μm coronal-projected sections showing the general distributions of EYFP-labeledCRH neurons in different brain regions. b EYFP-labeled clusters of fiber projections from CRH neurons in the AcbSh, IPAC, acp, cc, and icp. c EYFP-labeled single neurons in the CPu. d Clustered EYFP-labeled neurons distributed in the BAC. e EYFP-labeled neurons in the SCN. f Dense EYFP-labeled dendrites from neurons in the DC. g EYFP-labeled single neuron in the cerebellum. h EYFP-labeled fibers and swollen structures aroundthe 3V. i EYFP-labeled vascular-like structures in the VOLT. j EYFP-labeled terminals of PVN CRH neurons in the ME

Wang et al. BMC Biology (2021) 19:47 Page 5 of 18

shell (AcbSh), interstitial nucleus of the posterior limbof the anterior commissure (IPAC), anterior commis-sure, posterior (acp), corpus callosum (cc), and inferiorcerebellar peduncle (icp) (Fig. 2a, indicated by arrow-heads and Fig. 2b). The movies of serial sections showedthat the fibers in the AcbSh, IPAC, and acp were projec-tions from neurons in the OB (Additional file 2, Movie 1)and that fibers in the icp were projections from IO CRHneurons (Additional file 1, Figure S2. B and Add-itional file 3, Movie 2). Moreover, we found novel popu-lations of CRH-positive neurons in some brain regions,such as a sparsely distributed group in the CPu (Fig. 2c)that had dendrites that were radially distributed (withthe maximum radius from the terminals to the somatabeing 40–70 μm). Neurons gathered in the BAC (Fig. 2d)had round cell bodies and two short processes. Theaverage number of CRH-positive neurons found in theSCN was 45.67 ± 0.88, and nearly every neuron had twothick primary dendrites with few branches (Fig. 2e).Neurons in the dorsal cochlear nucleus (DC) (Fig. 2f)had dense apical dendrites distributed in the superficialglial zone, and chandelier cells with apical dendrites ver-tically distributed were labeled in the cerebellum(Fig. 2g). A cluster of swollen structures (Fig. 2h, indi-cated by arrowheads) presenting transparent smoothsurfaces was visible around the third ventricle (3 V) andalways extended to the 3 V border. Densely labeledvascular-like structures and terminals of CRH neuronswere found in VOLT (Fig. 2i) and ME (Fig. 2j). The dis-tributions and morphologies of CRH neurons in otherbrain regions are shown in Additional file 1, Figure S2.B.

Three-dimensional distributions and single-cellreconstructions of CRH neurons in several brain regionsWe reconstructed EYFP-labeled CRH neurons in severalbrain regions (Fig. 3a, h), including the OB (Fig. 3a), dor-sal part of lateral septal nucleus (LSD) (Fig. 3b), BST(Fig. 3c), CeA (Fig. 3d), VMPO (Fig. 3e), hippocampus(Hip) (Fig. 3f), SCN (Fig. 3g), and DC (Fig. 3h). Wefound that the reconstructed neurons in several brain re-gions (e.g., mPFC, BST, VMPO, anterior parvicellularpart of paraventricular hypothalamic nucleus (PaAp),periventricular hypothalamic nucleus (Pe), and SCN)shared similar morphological characteristics consistentwith bipolar neurons (Fig. 3i). CRH neurons in the LSDhad the largest average volume of somata (1632 ±159.6 μm3) (Fig. 3j) and the longest dendritic length(1.9 ± 0.2 mm) (Fig. 3k). Dendritic length significantly in-creased as a function of somatic volume (R2 = 0.597, P =0.0032) (Fig. 3n). CRH neurons in the VMPO also had alarger cell bodies (1172 ± 228.1 μm3) (Fig. 3j), but thenumber of dendritic branches (15.2 ± 3.7) (Fig. 3l) anddendritic length (1.1 ± 0.1 mm) (Fig. 3k) was less than

those of neurons in the LSD. Dendritic length was alsopositively correlated with somatic volume in the VMPO(Fig. 3o). Sholl analysis showed that neurons in the LSDhad the largest maximum number of intersections, whilethe VMPO had the least maximum number of intersec-tions. For all of these regions, the maximum numbers ofintersections were located at radial distances of 50–100 μm from the somata (Fig. 3m). The more complexdendrites of CRH neurons in the LSD, compared tothose in other areas, suggested that CRH neurons in theLSD may receive comparatively more inputs. Most ofthe dendritic morphologies of CRH neurons in thehippocampus exhibited a similar pattern of an umbrellashape of upward dendrites (Fig. 3f). CRH neurons in theSCN were scattered throughout the nucleus and thedendrites were interlaced with one another (Fig. 3g). Wenext compared the parameters of all reconstructed neu-rons (Additional file 1, Table S1) in different brain re-gions and found that CRH neurons in hypothalamicregions—including the PaAp (640.1 ± 60.4 μm3), Pe(951.2 ± 108.3 μm3), and SCN (636.0 ± 55.4 μm3)—hadsmaller somatic volumes (Fig. 3j and Additional file 1,Table S1). Similarly, there were also shorter dendriticlengths of CRH neurons in the PaAp (0.5 ± 0.02 mm), Pe(0.5 ± 0.06 mm), and SCN (0.6 ± 0.05 mm) (Fig. 3k andAdditional file 1, Table S1). The simpler morphologiesof hypothalamic CRH neurons may be related to theirendocrine and other conserved functions.

Multiple morphological types of CRH neurons formdistinct dendritic connections in the medial prefrontalcortex (mPFC)Recent studies have shown that CRH neurons in themPFC play a critical role in higher cognitive functions[29, 30]. In the prelimbic cortex (PrL) within the mPFC,we reconstructed the entire somata (Fig. 4A, purple bod-ies) and dendrites (Fig. 4A, color lines) of EYFP-labeledCRH neurons within a column that had a volume of350 × 500 × 500 μm. The cell bodies of these neuronswere mostly distributed within layers 2–4 and most oftheir dendrites were vertically distributed. There weredendritic branches in both the upper and lower parts ofthe somata. The apical dendrites that branched in thefirst layer formed a dense dendritic network, and mostof them reached the pia mater (Fig. 4A, reconstructed fi-bers indicated in layer 1); furthermore, the basal den-drites extended and branched into layer 4 at a distanceof approximately 500 μm from the cortical surface(Fig. 4A, reconstructed fibers indicated in layer 4). Indi-vidual reconstructed neurons were classified accordingto the distances (50–100, 100–150, 150–200, 200–250,and > 250 μm) between their somata and the surface ofthe cortex (Fig. 4B), and the percentages of neurons inthese categories were 19%, 31%, 35%, 10%, and 5%,

Wang et al. BMC Biology (2021) 19:47 Page 6 of 18

mPFC BST VMPO PaAp Pe SCN

0 25 50 7510012

515

017

520

022

525

027

530

032

535

037

540

0

0

5

10

15

Radial distance from soma ( µ m )

Nu

mb

ero

fin

ters

ecti

on

s LSD

VMPO

BST

0 1 2 30.0

0.5

1.0

1.5

2.0

2.5

Soma volume (x103µm3)

Den

dri

tic

len

gth

(mm

)

0 1 2 30

1

2

3

4

Soma volume (x103µ m3)

Den

dri

tic

len

gth

(mm

)

A B

C D

E F

G H

OB LSD

BST CeA

VMPO Hip

SCN DC

I

J K L

M

mPFC

LSD B

ST

VMPO HipCeA

PaAp Pe

SCN

0.0

0.5

1.0

1.5

2.0

So

ma

volu

me

(x10

3 µm

3 )

mPFC

LSDBST

VMPO Hip

CeAPaA

p PeSCN

0.0

0.5

1.0

1.5

2.0

2.5

To

talf

ibro

us

len

gth

( mm

)

N O

33 17 12 9 11 8 20 9 15 13 17 6 10 5 204 9 15

mPFC

LSD B

ST

VMPO Hip

CeAPaA

p PeSCN

0

10

20

30

40

50

60

Nu

mb

ero

ffi

bro

us

bra

nch

es

20

14 17 7 10 8 4

10 15

Fig. 3 (See legend on next page.)

Wang et al. BMC Biology (2021) 19:47 Page 7 of 18

(See figure on previous page.)Fig. 3 Three-dimensional distributions and single-cell reconstructions of CRH neurons in several brain regions. a–h Original images (left half) andreconstructions (right half) of CRH neurons in different brain regions, including the OB (a), LSD (b), BST (c), CeA (d), VMPO (e), Hip (f), SCN (g), andDC (h). The original images were inverted into grayscale images. The reconstructed somata and dendrites of neurons are indicated by purplebodies and red lines, respectively. i Typical reconstructed neurons show the common bipolar morphology found in different brain regions,including the mPFC, BST, VMPO, PaAp, Pe, and SCN. j Somatic volumes of CRH neurons in different brain regions [one-way ANOVA, P < 0.0001, F(8, 125) = 23.24], number in the bars indicate the number of neurons calculated. k Total fibrous lengths of CRH neurons in different brain regions[one-way ANOVA, P < 0.0001, F (8, 90) = 13.93]. l Numbers of fibrous branches of CRH neurons in different brain regions [one-way ANOVA, P <0.0001, F (8, 96) = 26.94]. m Sholl analysis of dendrites of neurons in the LSD, BST, and VMPO illustrate changes in the mean number ofintersections with increasing radial distance from the soma, n = 16 cells for LSD, 7 cells for BST and 10 cells for VMPO. n Correlation betweendendritic length and somatic volume in LSD neurons (Pearson’s correlation coefficient r = 0.68, P = 0.0029). o Correlation between dendritic lengthand somatic volume in VMPO neurons (Pearson’s correlation coefficient r = 0.88, P = 0.0019)

Fig. 4 (See legend on next page.)

Wang et al. BMC Biology (2021) 19:47 Page 8 of 18

respectively (Fig. 4B, indicated in the pie chart). Wefound that there was significant correlation between thesomatic depth (distance from the cortical surface) andboth the total dendritic length (Fig. 4C, upper half, r =0.4440) and total Euclidean distance (the straight-linedistance from the soma to the given point of the den-drite) (Fig. 4C, bottom half, r = 0.5399). Sholl analysesshowed that the number of intersections with a radialdistance from the soma being less than 50 μm was largerin neurons with a somatic depth of 50–100 μm than thatof neurons with a somatic depth of 100–150 μm; in con-trast, the number of intersections with a radial distancefrom the soma being more than 50 μm was smaller andended at approximately 100 μm of the radial distancefrom the soma in neurons with a somatic depth of 50–100 μm. Interestingly, the maximum numbers of inter-sections were similar between these two types of neu-rons (Fig. 4D). Sholl analyses of CRH neurons withdifferent somatic depths are shown in additional file 1,Figure S3. J. We also found that the total dendriticlength (Fig. 4E, left half, one-way ANOVA, P = 0.0008, F(2, 60) = 8.100) and the total Euclidean distance (Fig. 4E,right half, one-way ANOVA, P = 0.0002, F (2, 60) =9.915) of neurons with a somatic depth of less than100 μm were significantly smaller than those with som-atic depths of 100–150 μm (P = 0.0285) and more than150 μm (P = 0.0005), while there were no significant dif-ferences in the total number of dendritic branches orthe total number of dendritic terminal points (Additionalfile 1, Figure S3, K). The average dendritic lengths ofneurons at somatic depths of less than 100 μm, 100–150 μm, and more than 100–150 μm were 0.82 ± 0.26,

1.33 ± 0.54, and 1.48 ± 0.68 mm, respectively; further-more, their total Euclidean distances were 14.65 ± 5.19,38.77 ± 25.49, and 48.67 ± 30.38 mm, and their totalnumbers of dendritic branches were 23.23 ± 8.65,32.86 ± 20.28, 33.32 ± 19.44, respectively.To investigate local CRH-CRH connection patterns

within the cortex, we divided CRH-CRH connectionsinto three types (Fig. 4F, H). Type I consisted of basal-to-apical connections. Here, somata in layer 2 sent den-drites downward (the green cell of Fig. 4F, a) that con-tacted with the upward dendrites (the purple cell ofFig. 4F, a) from the somata in layer 3 (as shown in thered dotted box of Fig. 4F, a, b). Type II consisted ofbasal-to-somatic connections (as shown in the yellow-dotted box of Fig. 4Gc, d). In layer 2–3, a soma in theupper layer sent dendrites downward (as shown inFig. 4G, c, red cell), and the end of one branch (Fig. 4G,d red arrows) was in contact with an adjacent lower cellbody (Fig. 4G, c, orange cell; Fig. 4G, d, orange arrows).Type III consisted of basal-to-basal connections(Fig. 4G, c, e; Fig. 4H, f, green-dotted box). Two cellbodies in layers 2–3 sent dendrites downward, and theend of one branch (Fig. 4G, e, red arrows; Fig. 4H, g,yellow arrows) from the upper soma and the branch(Fig. 4G, e, orange arrows; Fig. 4H, g, blue arrows) fromthe lower soma formed a connection. A common fea-ture of the three types of connections was that thefluorescent intensity increased at the contact point, in-dicating a possible connection of structures (Fig. 4F, b,G, d and e, and H, g, purple arrows). Examples of type-II and type-III connections are demonstrated in Add-itional file 4, Movie 3.

(See figure on previous page.)Fig. 4 Multiple morphological types of CRH neurons form putative connections in the mPFC. A Overview of reconstructed CRH neuronsincluding somata (indicated by the purple bodies) and fibers (indicated by the colored lines) from layers 1–4 in a column with a volume of 350 ×500 × 500 μm in a PrL subregion within the mPFC. B Representation of 67 reconstructed somata and dendrites of CRH neurons arranged alongtheir somatic depths with respect to the pial surface; the pie chart at the bottom right shows the percentages of neurons distributed at differentsomatic depths. C Upper half: Correlation between total dendritic length and somatic depth in PrL neurons (Pearson’s correlation coefficient r =0.51, P < 0.0001); Lower half: Correlation between total Euclidean distance and somatic depth in PrL neurons (Pearson’s correlation coefficient r =0.59, P < 0.0001). D Sholl analysis of the dendrites of neurons with somatic depths = 50–100 μm and 100–150 μm illustrating changes in the meannumber of intersections with increasing radial distance from the soma. Inset images show exemplary intersections on two typical neurons (left:somatic depth = 50–100 μm, right: somatic depth = 100–150 μm) with different somatic depths, n = 15 cells for somatic depths = 50–100 μm and19 cells for somatic depths = 100–150 μm. E Left: The total dendritic length of CRH neurons with somatic depths < 100 μm showed differencescompared with those with somatic depths = 100–150 μm (one-way ANOVA, multiple comparisons, P = 0.0285) and soma depths > 150 μm (one-way ANOVA, multiple comparisons, P = 0.0005). Right: The total Euclidean distance of CRH neurons with a somatic depth of < 100 μm showeddifferences compared with those with soma depths = 100–150 μm (one-way ANOVA, multiple comparisons, P = 0.0213) and somatic depths >150 μm (one-way ANOVA, multiple comparisons, P = 0.0001), n = 13 cells for somatic depths < 100 μm, 21 cells for 100–150 μm and 31 cells for >150 μm. F–H Three types of connections between CRH neurons in the mPFC. F a (reconstructed neurons) and b (original image) show the type-Iconnection (the connection site is indicated by the purple arrowhead in the red dotted box). G The yellow-dotted boxes in c (reconstructedneurons) and d (original image) show the type-II connection (red arrowheads indicate a branch of basal dendrites of one neuron and the orangearrowhead indicates the cell body of another neuron; the connection site is indicated by the purple arrowhead). Green-dotted squares in c(reconstructed neurons) and e (original image) showed the type-III connection (red arrowheads indicated a branch of basal dendrites of oneneuron and the orange arrowheads indicate a branch of basal dendrites of another neuron; the connection site is indicated by the purplearrowhead). H Green-dotted squares in f (reconstructed neurons) and g (original image) also showed the type-III connection (yellow arrowheadsindicate a branch of basal dendrites of one neuron and the blue arrowheads indicate a branch of basal dendrites of another neuron; theconnection site is indicated by the purple arrowhead)

Wang et al. BMC Biology (2021) 19:47 Page 9 of 18

We next performed immunofluorescent staining to de-termine the specificity of EYFP-labeled neurons in CRH-IRES-Cre;Ai32 mice. The results showed that most ofthe EYFP-labeled neurons in the mPFC were CRH-immunoreactive cells (Additional file 1 Figure S3, A–C,indicated by white arrowheads). We further identifiedthat these CRH interneurons were GAD67-GFP-positiveneurons (Additional file 1, Figure S3, D-F, indicated bywhite arrowheads) by using CRH-IRES-Cre;Ai14;GAD67-GFP mice. Interestingly, in adult mouse brains,EYFP-labeled pyramidal neurons were visible in layer 3or layer 5 of the cortex (Additional file 1 Figure S3, H),but there were no EYFP-labeled pyramidal neurons onthe 21st day after birth (Additional file 1, Figure S3, G).These fluorescently labeled pyramidal neurons were notCRH-immunoreactive cells (Additional file 1, Figure S3,I), including within their dendrites and spines (Add-itional file 1, Figure S3, a, indicated by arrowheads). Wealso observed that some EYFP-labeled neurite swellingsin layer 1 were also labeled with CRH antibodies (Add-itional file 1, Figure S3, b, indicated by arrowheads).

Reconstructions and morphological features of CRHneurons in the PaAp and PeHypothalamic neuroendocrine CRH neurons play an im-portant role in stress responses, but neurons within dif-ferent subregions require more detailed morphologicalanalysis. We chose EYFP-labeled neurons in the PaAPand Pe to reconstruct their somata and processes(Fig. 5A, C; Additional file 1, Figure S4, A and B). Therewas a noteworthy co-localization pattern (Additional file1, Figure S4, E–G) for the EYFP-labeled signals andCRH immunoreactivity in the PaAP. There were vesicu-lar fluorescent labels (dendritic varicosities) (Fig. 5B, D,gray reconstructed structures) on the neurites of neu-rons in both the PaAP and Pe, and they also co-labeledwith CRH immunopositive-structures (Additional file 1,Figure S4, H, indicated by arrowheads). In terms of their3D patterns, the somata of some neurons (Fig. 5A, B,purple reconstructed cell bodies) distributed in the PaAPsent out fibers rostrally (Fig. 5A, B, red lines), and therewere spaced and small dendritic varicosities (Fig. 5B, b–d, gray bodies) on these fibers. An example of thesereconstructed cells is shown (Fig. 5B) according to theprimary branches and number of dendritic varicosities,and there were four distribution patterns of these neu-rons. Pattern 1 (Fig. 5B, a) consisted of cells that hadtwo primary branches with the shortest dendritic lengthand no varicosities. Pattern 2 (Fig. 5B, b) consisted ofcells that had two primary branches with similar den-dritic lengths at both ends of the somata and were dis-tributed almost vertically, and the fibers extendingrostrally had varicosities. Pattern 3 (Fig. 5B, c) consistedof cells with both ends of the dendrites having

varicosities and the one dendritic branch that was dis-tributed horizontally was longer and extended rostrally,whereas the other short branch was distributed verti-cally. Finally, pattern 4 (Fig. 5B, d) consisted of dendritesof multipolar cells extending rostrally having varicositiesand being distributed horizontally. The locations of thereconstructed somata in the PaAP are shown in Add-itional file 1, Figure S4 A (purple bodies indicated by redcircles).In the Pe, the reconstructed somata were located in

the lower part of the PaAp and around the 3V (Fig. 5C,purple bodies; Additional file 1, Figure S4, B, indicatedby red circles). These cells sent out fibers and one ortwo of them extended close to the 3V (Fig. 5C, D, redline). There were large varicosities on the dendrites(Fig. 5D, gray reconstituted bodies; Fig. 5E, indicated bywhite arrowheads) and they terminated (Fig. 5G, indi-cated by the yellow arrowhead) near the ependymal cell(Fig. 5G, indicated by the white arrowhead) layer adja-cent to the 3V. Most of the reconstructed cells withinthe Pe were bipolar neurons (Fig. 5D, e, f), and the fibersextending downward to the 3V had more varicosities(Fig. 5D, f, g). We further identified the immunopositivesubstances contained in these varicosities and found thatthere were extentive MAP 2 -immunopositive signals (amarker of dendrite) in the varicosities (Fig. 5E, indicatedby white arrowheads; Additional file 1, Figure S4, I).Chromogranin B (ChgB) immunoreactivity (associatedwith large dense core vesicles) was also found in thedendritic varicosities in PaAp (Fig. 5F) and Pe (Add-itional file 1, Figure S4, J-L) indicating that these varicos-ities contain a large amount of dense core vesicles.Interestingly, the ependymal cells adjacent to the 3Vwere also found to be ChgB immunopositive (Fig. 5G).The dendritic varicosities were also stained with Kine-sins (molecular motors used for intracellular transportand trafficking) in PaAp (Additional file 1, Figure S4, M-O) and Pe (Additional file 1, Figure S4, P-R).Next, we compared the somatic and dendritic parame-

ters of the neurons in the PaAP and Pe and found thatthe average volume of the soma in the Pe (951.2 ±108.3 μm3) was larger than that in the PaAP (640.1 ±60.4 μm3) (Fig. 5H, P = 0.0119, t = 2.697), and the num-ber of dendritic varicosities in the Pe (19.4 ± 2.5) was sig-nificantly greater than that in the PaAP (7.9 ± 1.6)(Fig. 5I, P = 0.0005, t = 3.974). The total dendriticvaricosities volume (2678 ± 652.3 μm3) (Fig. 5J, P = 0.0003,t = 4.348) per cell and the average volume of dendritic var-icosities (153.4 ± 43.1 μm3) (Fig. 5K, P = 0.0047, t = 3.148)in the Pe were significantly larger than those (468.2 ±79.2 μm3 and 47.7 ± 5.3 μm3) in the PaAP. We also foundthat there was a negative correlation between the numberof dendritic varicosities and the soma volume in the PaAP(r = − 0.4519, P = 0.0455) (Fig. 5L). There was no

Wang et al. BMC Biology (2021) 19:47 Page 10 of 18

Fig. 5 (See legend on next page.)

Wang et al. BMC Biology (2021) 19:47 Page 11 of 18

significant difference in the total dendritic length (PaAP:455.4 ± 24.1 μm, Pe: 537.5 ± 57.6 μm) (Fig. 5M) or the totalnumber of dendritic branches (PaAP: 5.4 ± 0.9, Pe: 6.3 ±1.6) (Fig. 5N) of neurons in the PaAP and Pe. We foundthat most (75% in the PaAP and 70% in the Pe) (Fig. 5O,right half, indicated in the pie chart) of the reconstructedneurons were bipolar neurons, which were characterizedby the number of primary dendritic branches (Fig. 5O, lefthalf).Collectively, these 3D reconstructions of hypothalamic

CRH neurons may be indicative of the transport andstorage of CRH peptides in hypothalamic neurons, aswell as the possible their release sites, such as the thirdventricle. These findings provide a structural basis forfurther elucidating the neural circuits and functions ofCRH neurons.

Arborization-dependent dendritic spine characteristics ofCRH neuronsWe further detected and analyzed the characteristics ofdendritic spines of CRH neurons. In general, CRH neu-rons with sparse dendritic branches had less spines.Consistent with previous studies, mushroom-like andthin dendritic spines were found in the cortex, hippo-campus, BST, and CeA (Fig. 6A, B). There were also sev-eral areas containing CRH neurons with dendritic spinesthat have not previously been reported. CRH neurons inthe VMPO and SCN were aspiny with few strongfilopodia-like spines (Fig. 6A, VMPO and SCN, indicatedby arrowheads), the maximum lengths of which reached5 μm. Furthermore, CRH neurons in the LSD hadmushroom-like spines (Fig. 6B, LSD, indicated by arrow-heads). CRH neurons with few dendritic branches

appeared to be aspiny, such as bipolar CRH neurons inthe cortex, VMPO, SCN (Fig. 6A), and BST (Fig. 6E),while CRH neurons with many branches (CRH neuronsshowed in Fig. 6B) were spiny. In the BST and CeA, wefurther calculated the densities of dendritic spines andfound that CRH neurons in the CeA had more spinesthan those in the BST (Fig. 6C, P < 0.0001, t = 5.467, 11different lengths of dendrites from three mice were cal-culated). Most oGAD67-GFP-positive CRH neurons inboth the BST (Fig. 6D, a) and CeA (Fig. 6F, b) werespiny, while aspiny CRH neurons (Fig. 6E) and GAD67-GFP-negative spiny CRH neurons (Fig. 6c) were alsofound in the BST and CeA. Interestingly, by injectingfluorogold into the mPFC (Fig. 6G) in CRH-IRES-Cre;Ai32 mice, we found that long-range-projecting CRHneurons that were co-labeled with fluorogold (Fig. 6H,d) in the anteromedial thalamic nucleus were aspiny(Fig. 6e, thin spines indicated by the arrows).

DiscussionDuring the last few decades, transgenic rodent modelshave become powerful tools for studying the distribution[15, 17, 19] and function [31–34] of CRH neurons in thebrain. In order to probe the morphological characteristicsof CRH neurons at single-cell resolution, we combinedgenetic labeling (using transgenic mouse lines) with thefMOST platform to generate high-resolution imagingdatasets, with which we characterized the morphologies ofdistinct CRH neurons distributed in various brain regionsthroughout the whole mouse brain.The robust native fluorescence of each of these re-

porter mouse lines enabled direct visualization of finedendritic and axonal structures of labeled neurons,

(See figure on previous page.)Fig. 5 Reconstructions and morphological features of CRH neurons in the PaAp and Pe. A Three-dimensional distribution of reconstructed wholemorphologies of CRH neurons in the PaAp. The somata are indicated by the purple bodies, the gray bodies indicate the dendritic varicosities, andthe red lines indicate the dendrites; the transparent structure on the left represents the reconstructed 3V. B Four typical reconstructed neurons inthe PaAp. B (a) A bipolar neuron with short dendrites and no varicosities; B (b) a bipolar neuron with varicosities located on the dendrite at oneend of the cell body. B (c) A bipolar neuron with varicosities located on dendrites at both ends of the cell body; B (d) a multipolar neuron withvaricosities located on the dendrites (scale bar = 100 μm). C Three-dimensional distribution of reconstructed whole morphologies of CRH neuronsin the Pe. The somata are indicated by purple bodies, the gray bodies indicate the dendritic varicosities, and the red lines indicate the dendrites;the transparent structure in the center represents the reconstructed 3V. D Three individual reconstructed neurons in the Pe. D (e) A bipolarneuron with both ends of dendrites extending to the 3V and with varicosities located on the dendrites at both ends of the cell body; D (f) abipolar neuron with one end of the dendrite extending to the 3V and with varicosities located on the dendrite at one end of the cell body; D (g)shows a multipolar neuron with one dendrite extending to the 3V and with varicosities mostly located on the dendrite at one end of the cellbody (scale bar = 100 μm). E EYFP-labeled varicosities contain MAP 2-immunopositive signals (indicated by the arrowheads, image below is amagnified image from the dotted box of the upper image, scale bar: 50 μm and 10 μm for the magnified image). F EYFP-labeled varicositiescontain ChgB-immunopositive signals (indicated by the arrowheads in the magnified images from the dotted boxes, scale bar: 50 μm and 10 μmfor the insert). G The dendrite extends to the third ventricle and a large varicosity (indicated by the yellow arrowhead) is attached to theependymal cells (indicated by the white arrowhead, scale bar: 10 μm). H–K Statistical results showing the differences of somatic volume (H)(P = 0.0119), varicosities number (I) (P = 0.0005), total varicosities volume per cell (J) (P = 0.0003), and the average volume of the varicosities (K)(P = 0.0047) between neurons in the PaAp and Pe. L Correlation between somatic volume and varicosities number of neurons in the PaAp.Spearman’s correlation coefficient was r = − 0.4519, P = 0.0455. Each point in the scatterplot represents a single cell. M, N There were nosignificant differences in dendritic features (M, total dendritic length; N, total number of dendritic branches) of neurons between PaAp and Pe.ns: not significant. O Comparison of primary dendritic branches and the proportions of bipolar or multipolar cells in the PaAp and Pe. N = 20 cellsfor PaAp and 10 cells for Pe

Wang et al. BMC Biology (2021) 19:47 Page 12 of 18

which has been demonstrated to be useful for mappingneuronal circuitry, as well as imaging and tracking spe-cific cell populations [35–37]. We compared the distri-bution patterns of fluorescent-labeled CRH neurons inthree reporter mouse lines. We found that adult CRH-IRES-Cre;Ai6 mice showed the highest number of la-beled neurons in several brain regions (Fig. 1J). Althoughthe three mouse lines were designed in a similar manner,the results may have been due to the sensitivity to Creand strength of fluorescent reporters. Ai6 reporter linesare more sensitive to low levels of Cre, leading to a morethorough identification of Cre-positive populations [35],and the expression of the enhanced fluorescent proteinZsGreen1 were more easily to be seen. Another possible

explanation is that Cre-mediated recombination had oc-curred in more cells in Ai14 or Ai32 reporter lines, butit was undetected owing to low reporter expression.Notably, the fluorescent fusion protein, CHR2-EYFP, ismembrane-bound and is therefore distributed along theplasma membrane of neuronal processes within CRH-IRES-Cre;Ai32 mice [38, 39], which enables a clearvisualization of the entire neuronal morphology. There-fore, we utilized CRH-IRES-Cre;Ai32 mice for whole-brain imaging and reconstructions. Interestingly, a largenumber of EYFP-labeled cortical pyramidal neurons wasalso observed in adult mice (which has not been re-ported previously from the onset age of postnatal day21) (Additional file 1, Figure S3).

Fig. 6 Dendritic spine characteristics of CRH neurons. A The dendritic spine characteristics of aspiny CRH neurons in brain regions of the cortex,VMPO, and SCN. The thin spines are indicated by the arrows and the filopodia-like spines are indicated by the yellow arrowheads. B The dendriticspine characteristics of spiny CRH neurons in brain regions of the LSD, hippocampus, BST, and CeA. The thin spines are indicated by the arrowsand the mushroom-like spines are indicated by the white arrowheads. The inserted images are magnified images from the dotted boxes (scalebar: 50 μm and 5 μm for the inserts). C The difference in dendritic spine density between the BST (n = 11) and CeA (n = 11) (P < 0.0001, t = 5.467).D (a) The GAD67-GFP-positive CRH neurons in the BST are spiny. a is a magnified image from the dotted box in D. E The aspiny CRH neurons inthe BST. F (b) and (c) The GAD67-GFP-positive (b) and -negative (c) CRH neurons in the CeA are spiny. b and c are magnified images from thedotted box in F. The inserted images are magnified images from the dotted boxes (scale bar: 50 μm for D and F, 10 μm for E and a–c. and 5 μmfor the inserts). G–H The retrograde-labeled CRH neurons in the AM (H, d) by injecting fluorogold in the mPFC (G) are aspiny (e, the thin spinesare indicated by the arrows). d and e are magnified images (scale bar: 1 mm for G and H, 50 μm for d, and 5 μm for e)

Wang et al. BMC Biology (2021) 19:47 Page 13 of 18

Next, we focused our analyses of reconstructed neu-rons mainly in several stress-related regions, includingthe mPFC, hypothalamus, amygdala, BST, and hippo-campus. For example, it has been reported that localCRH-synthesizing neurons are prominent in the PFC [8,40–43] and may modulate the activities of pyramidalneurons [7]. However, until now, the complete morph-ologies of CRH neurons in the mPFC have rarely beenreported. Here, we reconstructed fluorescent-labeledCRH neurons in the cortical column in the PrL withinmPFC across layers 1–4 (Fig. 4A). Importantly, we clas-sified different neuronal types by their soma depths andarborization patterns (example listed in Fig. 4B). For thefirst time, we showed the distribution of CRH neuronswith different morphological types in different corticallayers. We found there were dense dendrites (Fig. 4A)with dendritic swellings (Additional file 1, Figure S3, b)in layer 1 (the soma of which were located in layer 2/3or layer 4), and that fibers extended to the surface of thecortex. According to the layer-specific dendritic loca-tions and their different projection targets, several typesof putative connection patterns between CRH neuronswere identified in the cortex (listed in Fig. 4F–H). Sucha diverse dendritic connection pattern of cortical CRHneurons may reflect differential innervation of down-stream output targets (each amplified subfigure shownin Fig. 4F–H). Therefore, by characterizing their somaticlocations and unique respective local dendritic morph-ologies of CRH neurons, our present study not only in-creases our current understanding of the distribution ofCRH neurons, but also enables future studies to furtherelaborate upon cell-specific classifications. Taken to-gether, these findings may help elaborate future func-tional studies of morphologically diverse CRH neuronsin the PFC.Importantly, CRH functions as a neuropeptide hor-

mone produced in neuroendocrine neurons in the PVNand regulates the synthesis and secretion of glucocorti-coids from the adrenal glands through the action of ad-renocorticotropic hormone. For the first time, wereconstructed the intact morphologies of CRH neuronsand their neurite varicosities located within dendrites(Fig. 5E) in the PaAp (Fig. 5A, B) and Pe CRH neurons(Fig. 5C, D). In the PaAp, most dendrites with varicosi-ties projected rostrally (Fig. 5B), while in the Pe, the endsof the dendrites extended to the third ventricle and thelarge varicosities were attached to ependymal cells(Fig. 5C–D, G). We further identified that these den-dritic varicosities contained the large dense core vesicle-associated protein, ChgB, and molecular motors (e.g.,kinesins) used for intracellular transport and trafficking.Interestingly, a number of varicosities at the end of adendrite located closely to the out layer of ependymalcells to third ventricle which were ChgB immunopositive

(Fig. 5G). In addition, there were no EYFP-labeled CRHfibers distributed in the ependymal cells or passedthrough the cells. These results suggested that fibers ofCRH neurons in the Pe make direct contacts to epen-dymal cells and may release to the 3rd ventricle by epen-dymal cells. Therefore, we speculate that theseendocrine CRH neurons are different from those thatproject to the median eminence and that CRH may bealso released by dendrites to other areas of the hypothal-amus or cerebrospinal fluid to participate in its regula-tory functions. We further found a negative correlationbetween somatic volume and varicosities number in thePaAp (Fig. 5L). Thus, our reconstructed morphologicalcharacteristics of dendritic varicosities may facilitate fu-ture classifications (according to different fiber orienta-tions and varicosities distribution patterns) ofhypothalamic CRH neurons and advance our under-standing of their potentially diverse functions.The reconstructed CRH neurons in different brain re-

gions showed diverse distribution patterns and morph-ologies (Fig. 3a–h). We found that some neurons shareda common bipolar shape across various brain regions(Fig. 3i), especially in the hypothalamus (75% in thePaAp, 100% in SCN) and cortex. It has been reportedthat parvocellular CRH neuroendocrine neurons typic-ally have two relatively thick primary dendrites that ex-tend from opposite sides of the soma in a bipolararrangement and branch once [44, 45]; furthermore, bi-polar cells are commonly found in the cortex [12, 46,47]. Thus, our present study in CRH-reporter mice isconsistent with these previous studies and is the first todescribe the specific neural structures of these CRH neu-rons, such as the different types of connections betweenCRH neurons in the mPFC, as well as the intact morph-ologies of dendritic varicosities in hypothalamic CRHneurons. Such simplified branching properties of theseCRH neurons in the hypothalamus may be conducive totheir endocrine functions. Among all the reconstructedCRH neurons across different brain regions, theirsomata had different sizes (Fig. 3j), with somata in theLSD, VMPO, and hippocampus being larger than thosein the PaAp, Pe, and SCN. Also, we found differentialdendritic branch complexities across these regions. Forexample, CRH neurons in the mPFC, LSD, and hippo-campus all exhibited more complex dendritic morpholo-gies compared to those in the PaAP, Pe, and SCN(Fig. 3k–l). Hupalo et al. demonstrated that chemoge-netic activation of caudal but not rostral dmPFC CRHneurons potently impaired working memory, whereas in-hibition of these neurons improved working memory[29]. In addition, CRH acts in the medial septum to im-pair spatial memory [48] and acts in the BNST to par-ticipate in stress-induced maladaptive behaviors [49].However, the functions of CRH in the OB, SCN, and

Wang et al. BMC Biology (2021) 19:47 Page 14 of 18

VMPO remain unclear. Therefore, our current studymay provide a detailed morphological basis for futurefunctional-based studies on CRH neurons in these dif-ferent brain regions. Interestingly, the PaAP, Pe, andSCN are all contained within the hypothalamus, andCRH neurons in these regions had smaller somata andexhibited more prevalent bipolar branching patternscompared to those of other brain regions analyzed inour present study. In the PVN, CRH neurons have beenidentified as parvocellular cells [44, 50]. We found thatthe mean somatic volume of CRH neurons in the Pe waslarger than that of CRH neurons in the PaAP (Fig. 5H).Hence, we speculate that these data may be indicative oftwo different types of CRH endocrine neurons withinthe hypothalamus. Collectively, our quantitative analysisof these reconstructions demonstrates a region-specificdiversity of CRH neurons in terms of both somatic sizeand branching complexity.Dendritic spines are conventionally believed to be

largely absent from inhibitory neurons. Previous studiesby other groups and our previous research have shownthat CRH neurons are GABAergic neurons that are lo-cated in many different brain regions, such as in the cor-tex [30] and hippocampus; furthermore, CRH neuronsare usually aspiny, while some long-range projectingCRH neurons in the BST and CeA have been reportedto have spines [38]. In our present study, thearborization-dependent pattern of dendritic spines ofCRH neurons was detected where the most sophisticatedtypes of spines in the extended amygdala (BST and CeA)and the simplest one in the hypothalamus (VMPO andSCN). Interestingly, while spiny GABAergic CRH neu-rons in the BST and CeA were confirmed, aspiny CRHneurons were also found in these areas.

ConclusionsIn summary, in the present study, we generated high-resolution imaging datasets to characterize, at single-cellresolution, the fine morphologies of CRH neurons dis-tributed in diverse brain regions. Such region-specific re-constructions of intact morphologies of CRH neuronsmay help in further elucidating both CRH-mediatedphysiological functions in various brain circuits and theassociations of their dysfunction in various neuropatho-logical diseases.

MethodsAnimalsCRH-IRES-Cre (B6(Cg)-Crhtm1(cre)Zjh/J; stock number:012704), Ai6 (B6.Cg-Gt (ROSA)26Sortm6(CAG-Zs-

Green1)Hze/J; stock number: 007906), Ai14 (B6.Cg-Gt(ROSA)26Sortm14(CAG-TdTomato)Hze/J; stock number:007914), Gad67-GFP, and Ai32 (B6;Cg-Gt (ROSA)26-Sortm32(CAG-COP4*H134R/EYFP)Hze/J; stock number: 012569)

mice have been described previously [35, 36, 51, 52].CRH-IRES-Cre, Ai6, and Ai32 mice were purchasedfrom Jackson Laboratory. Gad67-GFP and Ai14 micewere obtained from the laboratories of Fuqiang Xu(WIPM, China) and Minmin Luo (NIBS, China), re-spectively. All of the mice were bred onto a C57BL/6 Jgenetic background. CRH-IRES-Cre;Ai6, CRH-IRES-Cre;Ai14, and CRH-IRES-Cre;Ai32 mice were derived fromcrosses of CRH-IRES-Cre/Ai6, Ai14, and Ai32 geno-types, respectively. CRH-IRES-Cre;Ai14;Gad67-GFPmice were derived from crosses of CRH-IRES-Cre;Ai14and Gad67-GFP genotypes. Male mice at 8–12 weeks ofage were used for experiments. Each group of 3–4 micewas used for visualizing and quantifying fluorescent-labeled neurons in three different mouse lines and den-dritic spines analysis. Three CRH-IRES-Cre;Ai32 malemice were used for fMOST imaging and neuronal recon-structions. The mice were housed on a 12-h light/darkcycle with food and water provided ad libitum. All ani-mal experiments were performed according to the pro-cedures approved by the Institutional Animal EthicsCommittee of the University of Science and Technologyof China.

HistologyAll histological procedures have been previously de-scribed [26, 53, 54]. Briefly, for whole-brain imaging,mice were anesthetized and perfused with 0.01M ofphosphate-buffered saline (PBS; Sigma-Aldrich Inc., St.Louis, USA), followed by 4% paraformaldehyde (PFA)and 2.5% sucrose in 0.01M of PBS. The brains were ex-cised and post-fixed in 4% PFA for 24 h. After fixation,each intact brain was rinsed overnight at 4 °C in 0.01Mof PBS and was subsequently dehydrated in a gradedethanol series. Then, the brains were impregnated withglycol methacrylate (GMA, Ted Pella Inc., Redding, CA)and embedded in a vacuum oven.For immunofluorescence and visualizing fluorescent-

labeled neurons in three different mouse lines, the fixedbrains were embedded by agarose and consecutive 50-or 100-μm-thick coronal sections were collected using avibrating microtome (Leica VT1200S, Germany). For im-munofluorescence, 50-μm-thick sections were washedthree times in PBS (10 min each time) and permeabilizedwith 0.3% Triton X-100 for 30 min, followed by incuba-tion in 5% normal-donkey-serum blocking solution atroom temperature for 1 h. Sections were then incubatedwith rabbit anti-CRF (1:2000, Bachem, T4037), rabbitanti-MAP 2 (1:200, SYSY, 188002), rabbit anti-Chromogranin B (1:100, Abcam, ab12242), or mouseanti-Kinesin (1:200, Millipore, MAB1614) primary anti-body in PBS containing 0.3% Triton X-100 overnight orfor 24–36 h at 4 °C. After washing in PBS, sections wereincubated with Alexa Fluor 594 or 647-conjugated

Wang et al. BMC Biology (2021) 19:47 Page 15 of 18

donkey anti-rabbit or donkey anti-mouse secondary anti-body (1:200, Jackson Immuno Research) diluted in 0.1%Triton X-100 in PBS at room temperature for 2 h. Afterwashing in PBS, sections were mounted on slides withantifade mounting medium (Vector Laboratories, Inc.,H-1000) and stored at 4 °C. For CRH immunofluores-cence, the mice were colchicine pretreated by intra-cerebroventricular injection of colchicine (0.2 mg/kg)and the samples were collected after 48 h. For visualizingfluorescent-labeled neurons in three different types ofmouse lines and dendritic spines, 100-μm-thick sectionswere washed in PBS and mounted on slides. All imageswere photographed using an LSM 880 (Zeiss, Germany)or FV3000 (Olympus, Japan) confocal microscope. Ab-breviations of brain regions are summarized in Add-itional file 1, Table S2.

Whole-brain imagingWhole-brain imaging was performed by the fMOST sys-tem [26]. Briefly, the immersed samples were fixed onthe imaging plane, and a WVT system automaticallyperformed the sectioning and imaging to complete thebrain-wide data acquisition. We acquired the data setsafter sectioning at a 1-μm thickness and imaging at avoxel size of 0.2 × 0.2 × 1 μm or 0.32 × 0.32 × 1 μm. Toenhance the in-focus EYFP signal, we added Na2CO3

into the water bath. Most of the EYFP molecules werepreserved in a nonfluorescent state, rather than directlydamaged, through chromophore protonation during theresin-embedding procedure. These fluorescent signalswere chemically recovered to the fluorescent state using0.05M of Na2CO3 during imaging. For the CRH-IRES-Cre;Ai32 samples, real-time PI staining was performed.

Image preprocessingThe raw data acquired by the fMOST system requiredimage preprocessing for mosaic stitching and illuminationcorrection. This process has been described previously[26]. Briefly, the mosaics of each coronal section werestitched to obtain an entire section based on accuratespatial orientation and adjacent overlap. Lateral illumin-ation correction was performed section by section. Imagepreprocessing was implemented in C++ and optimized inparallel using the Intel MPI Library (v.3.2.2.006, Intel).The whole data sets were executed on a computing server(72 cores, 2GHz per core) within 6 h. All full coronal sec-tions were saved at an 8-bit depth in LZW compressionTIFF format after image preprocessing.

Visualization and reconstructionWe visualized data sets using Amira software (v.5.2.2,FEI) and Imaris software (v.9.2.1, bitplane, Switzerland)to generate figures and movies. To process the TB-sizeddata on a single workstation, we transformed the data

format from TIFF to the native LDA type using Amira.The visualization process included extracting the data inthe range of interest, sampling or interpolation, reslicingthe images, identifying the maximum intensity projec-tion, volume and surface rendering, and generatingmovies using the main module of Amira. The segmenta-tion editor module of Amira was utilized for the manualoutline segmentation of the third ventricle, somata, andvaricosities. We applied the filament editor module ofAmira to trace the morphologies of EYFP-labeled neu-rons in 3D via a human-machine interaction. The recon-structed neurons were checked back-to-back by threeindividuals. The tracing results with original position in-formation were saved in SWC format and the results ofthe analyses were generated by L-Measure software(v.5.3).

StatisticsAll statistical graphs were generated using GraphPadPrism v.6.01. Two-tailed Student’s t tests and one-wayANOVAs, followed by Tukey’s post hoc tests, were alsoperformed using Graphpad Prism v.6.01 and SPSS (IBMSPSS Statistics 23). A P < 0.05 was considered to be sta-tistically significant. All results are presented as themean ± standard error of the mean (SEM).

Supplementary InformationThe online version contains supplementary material available at https://doi.org/10.1186/s12915-021-00973-x.

Additional file 1: Figure S1. Generation of transgenic mouse lines andmorphological features of CRH neurons in the BST and CeA. Figure S2.The novel EYFP-labeled CRH neurons identification and high-resolutionimages showing diverse morphologies of CRH neurons throughout thebrains of CRH-IRES-Cre;Ai32 mice. Figure S3. Expression specificity anddendritic analysis in the mPFC of CRH-IRES-Cre;Ai32 mice. Figure S4. Thesomatic locations and examples of the reconstructions of somata anddendritic varicosities of reconstructed neurons in the PaAp and Pe andimmunofluorescent staining identification of EYFP-labeled CRH neuronand the dendritic varicosities in the CRH-IRES-Cre;Ai32 mice. Table S1.Parameters of somatic volume, total dendritic length, and the number ofdendritic branches of the reconstructed neurons in several brain regions.Table S2. Abbreviation for brain regions.

Additional file 2: Movie 1. Movie of serial sections showing the fiberprojections from OB.

Additional file 3: Movie 2. Movie of serial sections showing the fiberprojections from IO.

Additional file 4: Movie 3. 3D movie showing the type II and type IIIconnections of CRH neurons in the mPFC.

Additional file 5. Excel file for the individual data values used in theFigs. 1 and 3 and the information of all antibodies used.

AcknowledgementsWe thank Dr. Fuqiang Xu (WIPM, China) and Dr. Minmin Luo (NIBS) forproviding the GAD67-GFP and Ai14 mice; Xiangning Li, Ren Miao, Ben Long,Jie Peng, Yuxin Li (HUST), Hui Fang, Xinya Qin, and Penghao Luo (USTC) forhelp with the experiment.

Wang et al. BMC Biology (2021) 19:47 Page 16 of 18

Authors’ contributionsH.G. and J.Z. designed research; Y.W., Q.S., C.H., P.C., and Z.H. performedresearch; A.L., and H.G. contributed new reagents/analytic tools; Y.W., Q.S.,A.L., H.G., and J.Z. analyzed data; and Y.W., P.H., and J.Z. wrote the paper. Allauthors read and approved the final manuscript.

FundingThis work was supported by the National Natural Science Foundation ofChina (Grant No. 32030046 and 81827901 and 32000716).

Availability of data and materialsAll data generated or analyzed during this study are included in this articleand its supplementary information files.

Ethics approval and consent to participateAll experimental procedures were approved by the Institutional AnimalEthics Committee of University of Science and Technology of China andwere performed humanely and in strict accordance with the InternationalGuiding Principles for Animal Research.

Consent for publicationNot applicable.

Competing interestsThe authors declare that they have no competing interests.

Author details1Chinese Academy of Science Key Laboratory of Brain Function and Diseases,School of Life Sciences, Division of Life Sciences and Medicine, University ofScience and Technology of China, Hefei 230026, China. 2Center forExcellence in Brain Science and Intelligence Technology, Chinese Academyof Sciences, Shanghai 200031, China. 3Britton Chance Center for BiomedicalPhotonics, Wuhan National Laboratory for Optoelectronics, MoE KeyLaboratory for Biomedical Photonics, School of Engineering Sciences,Huazhong University of Science and Technology, Wuhan 430074, China.

Received: 8 July 2020 Accepted: 1 February 2021

References1. Vale W, Spiess J, Rivier C, Rivier J. Characterization of a 41-residue ovine

hypothalamic peptide that stimulates secretion of corticotropin and beta-endorphin. Science. 1981;213(4514):1394–7.

2. Albeck DS, McKittrick CR, Blanchard DC, Blanchard RJ, Nikulina J, McEwenBS, Sakai RR. Chronic social stress alters levels of corticotropin-releasingfactor and arginine vasopressin mRNA in rat brain. J Neurosci. 1997;17(12):4895–903.

3. Mezey E, Kiss JZ, Skirboll LR, Goldstein M, Axelrod J. Increase ofcorticotropin-releasing factor staining in rat paraventricular nucleusneurones by depletion of hypothalamic adrenaline. Nature. 1984;310(5973):140–1.

4. Miyata M, Okada D, Hashimoto K, Kano M, Ito M. Corticotropin-releasingfactor plays a permissive role in cerebellar long-term depression. Neuron.1999;22(4):763–75.

5. Puder BA, Papka RE. Distribution and origin of corticotropin-releasing factor-immunoreactive axons in the female rat lumbosacral spinal cord. J NeurosciRes. 2001;66(6):1217–25.

6. Sasaki M, Sato H. Polysynaptic connections between Barrington’s nucleusand sacral preganglionic neurons. Neurosci Res. 2013;75(2):150–6.

7. Gallopin T, Geoffroy H, Rossier J, Lambolez B. Cortical sources of CRF, NKB,and CCK and their effects on pyramidal cells in the neocortex. Cereb Cortex.2006;16(10):1440–52.

8. Charlton BG, Ferrier IN, Perry RH. Distribution of corticotropin-releasingfactor-like immunoreactivity in human brain. Neuropeptides. 1987;10(4):329–34.

9. Bassett JL, Foote SL. Distribution of corticotropin-releasing factor-likeimmunoreactivity in squirrel monkey (Saimiri sciureus) amygdala. J CompNeurol. 1992;323(1):91–102.

10. Foote SL, Cha CI. Distribution of corticotropin-releasing-factor-likeimmunoreactivity in brainstem of two monkey species (Saimiri sciureus and

Macaca fascicularis): an immunohistochemical study. J Comp Neurol. 1988;276(2):239–64.

11. Merchenthaler I. Corticotropin releasing factor (CRF)-like immunoreactivity inthe rat central nervous system. Extrahypothalamic distribution. Peptides.1984;5(Suppl 1):53–69.

12. Delville Y, Stires C, Ferris CF. Distribution of corticotropin-releasing hormoneimmunoreactivity in golden hamster brain. Brain Res Bull. 1992;29(5):681–4.

13. Cummings SL. Distribution of corticotropin-releasing factor in thecerebellum and precerebellar nuclei of the cat. J Comp Neurol. 1989;289(4):657–75.

14. Shu YM, Ni RJ, Sun YJ, Fang H, Zhou JN. Distribution of corticotropin-releasing factor in the tree shrew brain. Brain Res. 2015;1618:270–85.

15. Wamsteeker Cusulin JI, Fuzesi T, Watts AG, Bains JS. Characterization ofcorticotropin-releasing hormone neurons in the paraventricular nucleus ofthe hypothalamus of Crh-IRES-Cre mutant mice. PLoS One. 2013;8(5):e64943.

16. Itoi K, Talukder AH, Fuse T, Kaneko T, Ozawa R, Sato T, Sugaya T, Uchida K,Yamazaki M, Abe M, Natsume R, Sakimura K. Visualization of corticotropin-releasing factor neurons by fluorescent proteins in the mouse brain andcharacterization of labeled neurons in the paraventricular nucleus of thehypothalamus. Endocrinology. 2014;155(10):4054–60.

17. Kono J, Konno K, Talukder AH, Fuse T, Abe M, Uchida K, Horio S, Sakimura K,Watanabe M, Itoi K. Distribution of corticotropin-releasing factor neurons inthe mouse brain: a study using corticotropin-releasing factor-modifiedyellow fluorescent protein knock-in mouse. Brain Struct Funct. 2017;222(4):1705–32.

18. Peng J, Long B, Yuan J, Peng X, Ni H, Li X, Gong H, Luo Q, Li A. Aquantitative analysis of the distribution of CRH neurons in whole mousebrain. Front Neuroanat. 2017;11:63.

19. Chen Y, Molet J, Gunn BG, Ressler K, Baram TZ. Diversity of reporterexpression patterns in transgenic mouse lines targeting corticotropin-releasing hormone-expressing neurons. Endocrinology. 2015;156(12):4769–80.

20. De Francesco PN, Valdivia S, Cabral A, Reynaldo M, Raingo J, Sakata I,Osborne-Lawrence S, Zigman JM, Perello M. Neuroanatomical andfunctional characterization of CRF neurons of the amygdala using a noveltransgenic mouse model. Neuroscience. 2015;289:153–65.

21. Huang L, Garcia I, Jen HI, Arenkiel BR. Reciprocal connectivity betweenmitral cells and external plexiform layer interneurons in the mouse olfactorybulb. Front Neural Circuits. 2013;7:32.

22. Garcia I, Bhullar PK, Tepe B, Ortiz-Guzman J, Huang L, Herman AM, ChaboubL, Deneen B, Justice NJ, Arenkiel BR. Local corticotropin releasing hormone(CRH) signals to its receptor CRHR1 during postnatal development of themouse olfactory bulb. Brain Struct Funct. 2016;221(1):1–20.

23. Nguyen AQ, Dela Cruz JA, Sun Y, Holmes TC, Xu X. Genetic cell targetinguncovers specific neuronal types and distinct subregions in the bed nucleusof the stria terminalis. J Comp Neurol. 2016;524(12):2379–99.

24. Ragan T, Kadiri LR, Venkataraju KU, Bahlmann K, Sutin J, Taranda J, Arganda-Carreras I, Kim Y, Seung HS, Osten P. Serial two-photon tomography forautomated ex vivo mouse brain imaging. Nat Methods. 2012;9(3):255–8.

25. Gong H, Zeng S, Yan C, Lv X, Yang Z, Xu T, Feng Z, Ding W, Qi X, Li A, Wu J,Luo Q. Continuously tracing brain-wide long-distance axonal projections inmice at a one-micron voxel resolution. Neuroimage. 2013;74:87–98.

26. Gong H, Xu D, Yuan J, Li X, Guo C, Peng J, Li Y, Schwarz LA, Li A, Hu B,Xiong B, Sun Q, Zhang Y, Liu J, Zhong Q, Xu T, Zeng S, Luo Q. High-throughput dual-colour precision imaging for brain-wide connectome withcytoarchitectonic landmarks at the cellular level. Nat Commun. 2016;7:12142.

27. Zheng T, Yang Z, Li A, Lv X, Zhou Z, Wang X, Qi X, Li S, Luo Q, Gong H,Zeng S. Visualization of brain circuits using two-photon fluorescence micro-optical sectioning tomography. Opt Express. 2013;21(8):9839–50.

28. Alon T, Zhou L, Perez CA, Garfield AS, Friedman JM, Heisler LK. Transgenicmice expressing green fluorescent protein under the control of thecorticotropin-releasing hormone promoter. Endocrinology. 2009;150(12):5626–32.

29. Hupalo S, Martin AJ, Green RK, Devilbiss DM, Berridge CW. Prefrontalcorticotropin-releasing factor (CRF) neurons act locally to modulatefrontostriatal cognition and circuit function. J Neurosci. 2019;39(11):2080–90.

30. Chen P, Lou S, Huang ZH, Wang Z, Shan QH, Wang Y, Yang Y, Li X, Gong H,Jin Y, Zhang Z, Zhou JN. Prefrontal cortex corticotropin-releasing factorneurons control behavioral style selection under challenging situations.Neuron. 2020;106(2):301–15. e7

Wang et al. BMC Biology (2021) 19:47 Page 17 of 18

31. Kolber BJ, Boyle MP, Wieczorek L, Kelley CL, Onwuzurike CC, Nettles SA,Vogt SK, Muglia LJ. Transient early-life forebrain corticotropin-releasinghormone elevation causes long-lasting anxiogenic and despair-like changesin mice. J Neurosci. 2010;30(7):2571–81.

32. Lu A, Steiner MA, Whittle N, Vogl AM, Walser SM, Ableitner M, Refojo D,Ekker M, Rubenstein JL, Stalla GK, Singewald N, Holsboer F, Wotjak CT, WurstW, Deussing JM. Conditional mouse mutants highlight mechanisms ofcorticotropin-releasing hormone effects on stress-coping behavior. MolPsychiatry. 2008;13(11):1028–42.

33. Regev L, Tsoory M, Gil S, Chen A. Site-specific genetic manipulation ofamygdala corticotropin-releasing factor reveals its imperative role in mediatingbehavioral response to challenge. Biol Psychiatry. 2012;71(4):317–26.

34. Stenzel-Poore MP, Heinrichs SC, Rivest S, Koob GF, Vale WW.Overproduction of corticotropin-releasing factor in transgenic mice: agenetic model of anxiogenic behavior. J Neurosci. 1994;14(5 Pt 1):2579–84.

35. Madisen L, Zwingman TA, Sunkin SM, Oh SW, Zariwala HA, Gu H, Ng LL,Palmiter RD, Hawrylycz MJ, Jones AR, Lein ES, Zeng H. A robust and high-throughput Cre reporting and characterization system for the whole mousebrain. Nat Neurosci. 2010;13(1):133–40.

36. Taniguchi H, He M, Wu P, Kim S, Paik R, Sugino K, Kvitsiani D, Fu Y, Lu J, LinY, Miyoshi G, Shima Y, Fishell G, Nelson SB, Huang ZJ. A resource of Credriver lines for genetic targeting of GABAergic neurons in cerebral cortex.Neuron. 2011;71(6):995–1013.

37. Walker LC, Cornish LC, Lawrence AJ, Campbell EJ. The effect of acute orrepeated stress on the corticotropin releasing factor system in the CRH-IRES-Cre mouse: a validation study. Neuropharmacology. 2019;154:96–106.

38. Dedic N, Kuhne C, Jakovcevski M, Hartmann J, Genewsky AJ, Gomes KS,Anderzhanova E, Pohlmann ML, Chang S, Kolarz A, Vogl AM, Dine J, MetzgerMW, Schmid B, Almada RC, Ressler KJ, Wotjak CT, Grinevich V, Chen A, SchmidtMV, Wurst W, Refojo D, Deussing JM. Chronic CRH depletion from GABAergic,long-range projection neurons in the extended amygdala reduces dopaminerelease and increases anxiety. Nat Neurosci. 2018;21(6):803–7.

39. Park SJH, Pottackal J, Ke JB, Jun NY, Rahmani P, Kim IJ, Singer JH, Demb JB.Convergence and divergence of CRH Amacrine cells in mouse retinalcircuitry. J Neurosci. 2018;38(15):3753–66.

40. Swanson LW, Sawchenko PE, Rivier J, Vale WW. Organization of ovinecorticotropin-releasing factor immunoreactive cells and fibers in the rat brain:an immunohistochemical study. Neuroendocrinology. 1983;36(3):165–86.

41. De Souza EB, Insel TR, Perrin MH, Rivier J, Vale WW, Kuhar MJ. Corticotropin-releasing factor receptors are widely distributed within the rat centralnervous system: an autoradiographic study. J Neurosci. 1985;5(12):3189–203.

42. Merchenthaler I, Hynes MA, Vigh S, Schally AV, Petrusz P. Corticotropin releasingfactor (CRF): origin and course of afferent pathways to the median eminence(ME) of the rat hypothalamus. Neuroendocrinology. 1984;39(4):296–306.

43. Lewis DA, Foote SL, Cha CI. Corticotropin-releasing factor immunoreactivityin monkey neocortex: an immunohistochemical analysis. J Comp Neurol.1989;290(4):599–613.

44. Rho JH, Swanson LW. A morphometric analysis of functionally definedsubpopulations of neurons in the paraventricular nucleus of the rat withobservations on the effects of colchicine. J Neurosci. 1989;9(4):1375–88.

45. Bittar TP, Nair BB, Kim JS, Chandrasekera D, Sherrington A, Iremonger KJ.Corticosterone mediated functional and structural plasticity in corticotropin-releasing hormone neurons. Neuropharmacology. 2019;154:79–86.

46. Cahusac PM, Castro MG, Robertson L, Lowenstein PR. Electrophysiologicalevidence against a neurotransmitter role of corticotropin-releasing hormone(CRH) in primary somatosensory cortex. Brain Res. 1998;793(1–2):73–8.

47. Yan XX, Baram TZ, Gerth A, Schultz L, Ribak CE. Co-localization of corticotropin-releasing hormone with glutamate decarboxylase and calcium-binding proteinsin infant rat neocortical interneurons. Exp Brain Res. 1998;123(3):334–40.

48. Hupalo S, Bryce CA, Bangasser DA, Berridge CW, Valentino RJ, Floresco SB.Corticotropin-releasing factor (CRF) circuit modulation of cognition andmotivation. Neurosci Biobehav Rev. 2019;103:50–9.

49. Hu P, Liu J, Maita I, Kwok C, Gu E, Gergues MM, Kelada F, Phan M, Zhou JN,Swaab DF, Pang Z, Lucassen PJ, Roepke TA, Samuels BA. Chronic stressinduces maladaptive behaviors by activating corticotropin-releasinghormone signaling in the mouse oval bed nucleus of the stria terminalis.J Neurosci. 2020;40(12):2519–37.

50. Miklos IH, Kovacs KJ. GABAergic innervation of corticotropin-releasinghormone (CRH)-secreting parvocellular neurons and its plasticity asdemonstrated by quantitative immunoelectron microscopy. Neuroscience.2002;113(3):581–92.

51. Madisen L, Mao T, Koch H, Zhuo JM, Berenyi A, Fujisawa S, Hsu YW, GarciaAJ 3rd, Gu X, Zanella S, Kidney J, Gu H, Mao Y, Hooks BM, Boyden ES,Buzsaki G, Ramirez JM, Jones AR, Svoboda K, Han X, Turner EE, Zeng H. Atoolbox of Cre-dependent optogenetic transgenic mice for light-inducedactivation and silencing. Nat Neurosci. 2012;15(5):793–802.

52. Tamamaki N, Yanagawa Y, Tomioka R, Miyazaki J, Obata K, Kaneko T. Greenfluorescent protein expression and colocalization with calretinin,parvalbumin, and somatostatin in the GAD67-GFP knock-in mouse. J CompNeurol. 2003;467(1):60–79.

53. Yang Z, Hu B, Zhang Y, Luo Q, Gong H. Development of a plasticembedding method for large-volume and fluorescent-protein-expressingtissues. PLoS One. 2013;8(4):e60877.

54. Xiong H, Zhou Z, Zhu M, Lv X, Li A, Li S, Li L, Yang T, Wang S, Yang Z, Xu T,Luo Q, Gong H, Zeng S. Chemical reactivation of quenched fluorescentprotein molecules enables resin-embedded fluorescence microimaging. NatCommun. 2014;5:3992.

Publisher’s NoteSpringer Nature remains neutral with regard to jurisdictional claims inpublished maps and institutional affiliations.

Wang et al. BMC Biology (2021) 19:47 Page 18 of 18